A novel information theoretic approach to wavelet feature selection for texture classification

In this research we address the problem of discriminant subband selection for texture classification. A novel Effective Information based Subband Selection (EISS) algorithm is proposed which utilizes the intra-class and inter-class distributions. Essentially these distributions are used to calculate the class-based entropy for a given subband. This class-based information is incorporated in the total information content of the training images to develop a robust Effective Information (EI) criterion. Only the subbands with the top EI criteria are allowed to participate in the classification process. The proposed EISS algorithm is evaluated on Brodatz texture database and has shown to outperform the most relevant method based on mutual information criterion.

Figure optionsDownload as PowerPoint slideHighlights
► In this research we address the problem of discriminant subband selection for efficient texture classification.
► A novel Effective Information criterion is proposed for robust subband selection.
► This is for the first time that Effective Information is used for subband selection.
► Extensive experiments have been conducted on synthetic and real databases showing the efficacy of the proposed approach compared to the previous state-of-art.